u-boot.lists.denx.de archive mirror
 help / color / mirror / Atom feed
From: Mark Kettenis <mark.kettenis@xs4all.nl>
To: AKASHI Takahiro <takahiro.akashi@linaro.org>
Cc: xypron.glpk@gmx.de, agraf@csgraf.de, sjg@chromium.org,
	ilias.apalodimas@linaro.org, u-boot@lists.denx.de,
	takahiro.akashi@linaro.org
Subject: Re: [RFC 2/3] efi_loader: bootmgr: add booting from removable media
Date: Tue, 9 Nov 2021 14:53:14 +0100 (CET)	[thread overview]
Message-ID: <d3cac2e5b37f96b5@bloch.sibelius.xs4all.nl> (raw)
In-Reply-To: <20211109013233.72902-3-takahiro.akashi@linaro.org> (message from AKASHI Takahiro on Tue, 9 Nov 2021 10:32:32 +0900)

> From: AKASHI Takahiro <takahiro.akashi@linaro.org>
> Date: Tue,  9 Nov 2021 10:32:32 +0900

Hi Takahiro,

> Under the current implementation, booting from removal media using
> a architecture-specific default image name, say BOOTAA64.EFI, is
> supported only in distro_bootcmd script. See the commit 74522c898b35
> ("efi_loader: Add distro boot script for removable media").
> 
> This is, however, half-baked implementation because
> 1) UEFI specification requires this feature to be implemented as part
>    of Boot Manager's responsibility:
> 
>   3 - Boot Manager
>   3.5.1 Boot via the Simple File Protocol
>   When booting via the EFI_SIMPLE_FILE_SYSTEM_PROTOCOL, the FilePath will
>   start with a device path that points to the device that implements the
>   EFI_SIMPLE_FILE_SYSTEM_PROTOCOL or the EFI_BLOCK_IO_PROTOCOL. The next
>   part of the FilePath may point to the file name, including
>   subdirectories, which contain the bootable image. If the file name is
>   a null device path, the file name must be generated from the rules
>   defined below.
>   ...
>   3.5.1.1 Removable Media Boot Behavior
>   To generate a file name when none is present in the FilePath, the
>   firmware must append a default file name in the form
>   \EFI\BOOT\BOOT{machine type short-name}.EFI ...
> 
> 2) So (1) entails the hehavior that the user's preference of boot media
>    order should be determined by Boot#### and BootOrder variables.
> 
> With this patch, the semantics mentioned above is fully implemented.
> For example, if you want to boot the system from USB and SCSI in this
> order,
> * define Boot0001 which contains only a device path to the USB device
>   (without any file path/name)
> * define Boot0002 which contains only a device path to the SCSI device,
> and
> * set BootOrder to Boot0001:Boot0002

The problem with this approach is that EFI device paths are
complicated things and I don't think we can expect users to do what
you describe above all by hand.  All x86 machines I've seen have a
more user-friendly way to configure the desired boot order, and also a
way to specify which device to boot from after automatic boot has been
interrupted.

With the current distroboot scripts one can simply do:

  setenv boot_targets usb0 mmc0 pxe dhcp

and I think we need to have something simple like that.  Maybe the
code can automatically create Boot#### and BootOrder variables based
on the boot_targets environment variables if they're not defined?

Another idea is to accept U-Boot device names (e.g. usb0, mmc0) as
aliases for EFI device paths and automatically expand them?

> Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
> ---
>  lib/efi_loader/efi_bootmgr.c | 65 +++++++++++++++++++++++++++++++++++-
>  1 file changed, 64 insertions(+), 1 deletion(-)
> 
> diff --git a/lib/efi_loader/efi_bootmgr.c b/lib/efi_loader/efi_bootmgr.c
> index 1fe19237f9a6..1d9d5858561f 100644
> --- a/lib/efi_loader/efi_bootmgr.c
> +++ b/lib/efi_loader/efi_bootmgr.c
> @@ -30,6 +30,66 @@ static const struct efi_runtime_services *rs;
>   * should do normal or recovery boot.
>   */
>  
> +#if defined(CONFIG_ARM64)
> +#define BOOTEFI_NAME "bootaa64.efi"
> +#elif defined(CONFIG_ARM)
> +#define BOOTEFI_NAME "bootarm.efi"
> +#elif defined(CONFIG_X86_RUN_32BIT)
> +#define BOOTEFI_NAME "bootia32.efi"
> +#elif defined(CONFIG_X86_RUN_64BIT)
> +#define BOOTEFI_NAME "bootx64.efi"
> +#elif defined(CONFIG_ARCH_RV32I)
> +#define BOOTEFI_NAME "bootriscv32.efi"
> +#elif defined(CONFIG_ARCH_RV64I)
> +#define BOOTEFI_NAME "bootriscv64.efi"
> +#else
> +#define BOOTEFI_NAME "dummy.efi"
> +#endif
> +
> +/**
> + * expand_media_path() - expand a device path for default file name
> + * @device_path:	device path to check against
> + *
> + * If @device_path is a media or disk partition which houses a file
> + * system, this function returns a full device path which contains
> + * an architecture-specific default file name for removable media.
> + *
> + * Return:	a newly allocated device path
> + */
> +static
> +struct efi_device_path *expand_media_path(struct efi_device_path *device_path)
> +{
> +	struct efi_device_path *dp, *full_path;
> +	efi_handle_t handle;
> +	efi_status_t ret;
> +
> +	if (!device_path)
> +		return NULL;
> +
> +	/*
> +	 * If device_path is a (removable) media or partition which provides
> +	 * simple file system protocol, append a default file name to support
> +	 * booting from removable media.
> +	 */
> +	dp = device_path;
> +	ret = efi_locate_device_path(&efi_simple_file_system_protocol_guid,
> +				     &dp, &handle);
> +	if (ret == EFI_SUCCESS) {
> +		if (dp->type == DEVICE_PATH_TYPE_END) {
> +			dp = efi_dp_from_file(NULL, 0,
> +					      "/efi/boot/" BOOTEFI_NAME);
> +			full_path = efi_dp_append(device_path, dp);
> +		} else {
> +			full_path = efi_dp_dup(device_path);
> +		}
> +		efi_free_pool(dp);
> +	} else {
> +		full_path = efi_dp_dup(device_path);
> +	}
> +
> +	return full_path;
> +}
> +
>  /**
>   * try_load_entry() - try to load image for boot option
>   *
> @@ -68,13 +128,16 @@ static efi_status_t try_load_entry(u16 n, efi_handle_t *handle,
>  	}
>  
>  	if (lo.attributes & LOAD_OPTION_ACTIVE) {
> +		struct efi_device_path *file_path;
>  		u32 attributes;
>  
>  		log_debug("%s: trying to load \"%ls\" from %pD\n",
>  			  __func__, lo.label, lo.file_path);
>  
> -		ret = EFI_CALL(efi_load_image(true, efi_root, lo.file_path,
> +		file_path = expand_media_path(lo.file_path);
> +		ret = EFI_CALL(efi_load_image(true, efi_root, file_path,
>  					      NULL, 0, handle));
> +		efi_free_pool(file_path);
>  		if (ret != EFI_SUCCESS) {
>  			log_warning("Loading %ls '%ls' failed\n",
>  				    varname, lo.label);
> -- 
> 2.33.0
> 
> 

  parent reply	other threads:[~2021-11-09 13:53 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-09  1:32 [RFC 0/3] efi_loader: bootmgr itself should support removable media AKASHI Takahiro
2021-11-09  1:32 ` [RFC 1/3] efi_loader: export efi_locate_device_handle() AKASHI Takahiro
2021-11-09  1:32 ` [RFC 2/3] efi_loader: bootmgr: add booting from removable media AKASHI Takahiro
2021-11-09  9:14   ` Heinrich Schuchardt
2021-11-10  5:38     ` AKASHI Takahiro
2021-11-09 13:53   ` Mark Kettenis [this message]
2021-11-10  5:52     ` AKASHI Takahiro
2021-11-09  1:32 ` [RFC 3/3] cmd: efidebug: handle " AKASHI Takahiro
2021-11-09  9:50   ` Heinrich Schuchardt
2021-11-10  6:24     ` AKASHI Takahiro
2021-11-10  8:11       ` Heinrich Schuchardt
2021-11-10  9:17         ` Heinrich Schuchardt
2021-11-12  1:51           ` AKASHI Takahiro

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d3cac2e5b37f96b5@bloch.sibelius.xs4all.nl \
    --to=mark.kettenis@xs4all.nl \
    --cc=agraf@csgraf.de \
    --cc=ilias.apalodimas@linaro.org \
    --cc=sjg@chromium.org \
    --cc=takahiro.akashi@linaro.org \
    --cc=u-boot@lists.denx.de \
    --cc=xypron.glpk@gmx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).